Embodiments of the present invention generally relates to the field of data processing, and more specifically, to a method and system for processing semantic fragments.
In most cases, data need to be processed according to predetermined steps or phases to accomplish a particular data processing task. These steps or phases for data processing are generally specified by a data processing flow. Specifically, the data processing flow may specify nodes for processing data, operations or activities performed for the data at these nodes, and data flow between the nodes, etc. The data processing flow may be parsed by a data processing engine for controlling data distribution and processing. For example, for traffic implemented in a digitalized manner, data related to the traffic may be processed hierarchically in a predetermined sequence between different personnel, departments and/or branches. A flow describing such data processing is called “traffic flow,” which is one of examples of a data processing flow.
In order to complete different data processing tasks, it may be required to implement different data processing flows. Therefore, the amount of data processing flows could be huge. For example, there could be thousands of traffic flows associated with different traffics only in one mechanism. Moreover, many traffic flows may be heterogeneous, i.e., they involve different data processing nodes, operations and/or data flow paths. The huge amount and heterogeneity of the data processing flows may increase the complexity and load of data processing engines, which may affect the efficiency of data processing. Degradation of the data processing efficiency may in turn prolong the response time, which dampens user experience. Moreover, the maintenance difficulty associated with data processing flows and data processing engine may also increase.
In view of the above, there is a need in the art for a method of processing semantic fragments to merge a data processing flow in an appropriate manner.